[PATCH v2 01/15] RISC-V: Build Infastructure

[Palmer Dabbelt]

On Wed, 20 Dec 2017 06:05:20 PST (-0800), ldv@altlinux.org wrote:
> On Tue, Dec 19, 2017 at 11:20:08PM -0800, Palmer Dabbelt wrote:
>> This patch lays out the top-level orginazition of the RISC-V port.  It
>> contains all the Implies files as well as various other fragments of
>> build infastructure for the RISC-V port.  This contains the only change
>> to a shared file: config.h.in.
> [...]
>>  config.h.in                                  |   5 +
>>  nptl/Makefile                                |   4 +
>
> Well, nptl/Makefile is a shared file, too.
>
> [...]
>> diff --git a/nptl/Makefile b/nptl/Makefile

>> index 60d036a1ea42..f62d19febf1d 100644
>> --- a/nptl/Makefile
>> +++ b/nptl/Makefile
>> @@ -611,6 +611,10 @@ else
>>  librt = $(common-objpfx)rt/librt.a
>>  endif
>>  
>> +# `make check' sometimes triggers a rebuild of librt.so using this Makefile,
>> +# which ignores librt's dependence on libpthread
>> +$(common-objpfx)rt/librt.so: $(shared-thread-library)
>> +
>>  $(objpfx)tst-cancel17: $(librt)
>>  $(objpfx)tst-cancelx17: $(librt)
>>  $(objpfx)tst-_res1mod2.so: $(objpfx)tst-_res1mod1.so
>
> In fact, this looks like a fix unrelated to the RISC-V port.
> If this is really the case, please submit the fix separately.
>
> Thanks,

Oh, sorry, that must have snuck in somehow after I wrote the commit message -- 
I did a lot of rebasing.  We noticed this bug when running the test suite, I 
think Andrew must have fixed it because I was still using a workaround.

Thanks for catching this, I'll submit a patch that's pulled out from our port.